AI Solution Architect

Roles & Responsibilities : This role architects production-ready AI and GenAI solutions that are reliable, observable, safe, and cost-efficient. The architect owns the technical design of AI solutions from opportunity through delivery. The solution architect is the senior technical authority on AI architecture decisions within projects and collaborates with the Data Solution Architect where engagements span both data foundations and AI solutions. Key responsibilities AI & GenAI solution architecture Own end-to-end architecture for AI and GenAI engagements — classical ML, LLM applications, RAG systems, and agentic workflows. Responsible for design decisions: build vs. buy, model selection, fine-tuning vs. retrieval, and orchestration patterns. Design retrieval architectures (chunking, embedding, indexing, hybrid search, and re-ranking) for accuracy and latency at scale. Architect agentic systems: tool use, memory, multi-step orchestration, and human-in-the-loop control points. Production readiness & LLMOps design Design LLMOps/MLOps setup for engagements: model serving, versioning, deployment pipelines, and rollback. Define evaluation architecture: eval harnesses, quality benchmarks, regression testing, and acceptance criteria. Architect observability, monitoring, and drift detection; design for operability and handoff to Managed Ops. Design cost-efficient inference architectures: model tiering, caching, token economics, and FinOps guardrails. Data foundations for AI Specify the data requirements for AI solutions — training data, feature pipelines etc. Partner with Data Solution Architects to translate model and RAG requirements into data platform design. Desig